Author: AIImpish
-

PlanAssist Goes to 15 English Councils: How Google DeepMind’s Gemini Tool Speeds Up Home Planning Checks
PlanAssist, built with Gemini, moves from three pilot councils to a 12-month beta in 15 English councils. What it does, the claimed time savings, and the limits.
-

GPT-Rosalind-6: OpenAI’s Life-Sciences Model Now Runs on GPT-6 Astra, but Only for Vetted Researchers
GPT-Rosalind-6 uses GPT-6 Astra’s weights, scores higher on OpenAI’s science benchmarks, and is only available to vetted researchers. Here is who can use it.
-

Microsoft’s Mage-VL-Cowork: A 4B Open Model That Knows When to Click and When to Code
Microsoft’s open 4B Mage-VL-Cowork switches between clicking and running code, with a reported 66.0% on OSWorld-Verified. What it does and how to test it safely.
-

Apple Intelligence Is Coming to China: Testing Starts November 4, With Alibaba’s Qwen Inside Siri
Apple Intelligence testing starts in China on November 4, with a public iOS 27.3 release in December and Alibaba’s Qwen inside Siri and Writing Tools.
-

ChatGPT Tightens Weapons Limits for Teens and Past-Attack Details for Everyone: What Parents Should Know
OpenAI is limiting weapons information for teens, reducing detail on past mass attacks for everyone, and promising a twice-yearly Community Safety Report.
-

One robots.txt Rule to Opt Out of AI Training: What Google, Microsoft, OpenAI, and Anthropic Agreed To
Google, Microsoft, OpenAI, and Anthropic back one shared robots.txt token to opt out of AI training. Here is what it does, when it takes effect, and how to add it.
-

Genie 4 explained: Google’s world model that generates playable scenes with sound, and what it still can’t do
Google DeepMind announced Genie 4 on October 27, 2026 — a new world model for real-time interactive simulation with native audio. You steer a scene with keyboard controls and text prompts; Genie 4 simulates what happens next, including how the world looks, how it sounds, and how people in it move and speak. It is…
-

ElevenLabs SFX v3: multi-sound prompts for creators, plus a prompt pack for video and game sound
ElevenLabs launched SFX v3 on October 29, 2026 — a new text-to-sound-effects model aimed at scenes that need more than one sound in a single clip. If you make short video, game audio, or social Foley, this is the update that matters: one prompt can ask for a crowd, a ball drop, and a boo…
-

Gemini API now stores your requests by default on new paid projects: how to check and turn it off
Google changed a quiet but important default on October 26, 2026. If you create a new paid-tier Gemini API project and call generateContent, streamGenerateContent, or batchGenerateContent, those requests are now stored by default (store=true). That matches how the Interactions API already behaved. Older projects keep whatever setting they already had. This is a privacy and…
-

ChatGPT Pro Gets Double the Work and Codex Usage, Plus Free Monthly API Credits: What Changed and Who Benefits
On October 29, 2026, OpenAI changed what its top ChatGPT plans include. Pro 100 and Pro 500 now get twice their previous Work and Codex allowance, a planned Pro 200 cut was canceled, and eligible Pro and Business subscribers can now claim monthly API credits to build their own apps. Coffee summary FACT: Pro 100…